The Research Process

1. Formulating the Research Problem

• The first step in research is to identify and formulate the research problem.
• Problems may relate either to states of nature or to relationships between variables.
• At the outset, the researcher must single out the problem he wants to study, decide
the general area of interest, and narrow it down.
• The problem may be stated initially in broad terms, but ambiguities relating to it should
be resolved.
• Feasibility of possible solutions must be examined before setting up a working
formulation.
• Two important tasks are involved: (i) understanding the problem thoroughly, and (ii)
rephrasing it into meaningful terms from an analytical point of view.
• The best way of understanding the problem is to discuss it with colleagues, experts, or
guides in academic institutions.
• Reviewing the available literature on the subject is also important in this step.
• A well-defined problem statement is of great importance because it determines the
type of data, techniques, and report to be prepared.
2. Extensive Literature Survey

• Once the problem is formulated, the researcher should write down a brief summary of
it.
• At this stage, a synopsis is often required (e.g., in Ph.D. work) to be approved by a
research board or committee.
• The researcher must conduct an extensive survey of the available literature related to
the problem.
• Useful sources include abstracting and indexing journals, published and unpublished
bibliographies, government reports, and academic journals.
• Books, research papers, and conference proceedings must also be studied depending
upon the nature of the problem.
• During the survey, one source of information usually leads to another.
• Earlier studies, if similar to the one in hand, should be carefully examined.
• A good library is a very helpful resource at this stage.
• The outcome of the survey is that the researcher becomes well acquainted with the
problem and available data.

3. Developing the Hypothesis

• After the literature survey, the researcher should clearly state the working hypothesis
or hypotheses.
• A working hypothesis is a tentative assumption made to draw out and test logical or
empirical consequences.
• Hypotheses provide the focal point for research and affect the type of tests and data
analysis to be used.
• They must be specific and limited to the piece of research in hand.
• Hypotheses guide the researcher by delimiting the scope and keeping him on the right
track.
• They sharpen thinking, focus attention on important aspects, and indicate the type of
data and analysis required.
• Hypotheses may be developed through discussions with experts, examination of data,
review of similar studies, or exploratory investigations.
• They may also arise from a priori thinking about the subject or counsel of experienced
people.
• Occasionally, in exploratory or formulative research, hypotheses may not be required.
4. Preparing the Research Design

• Once the problem and hypotheses are formulated, the researcher prepares the
research design.
• Research design is the conceptual structure within which research is conducted.
• It functions as the blueprint for the collection, measurement, and analysis of data.
• A good design makes research efficient, yielding maximum information with minimal
cost, time, and effort.
• The purpose of research (exploration, description, diagnosis, or experimentation)
largely determines the design.
• For exploration, a flexible design is considered appropriate, whereas descriptive or
diagnostic studies need designs that minimize bias and maximize reliability.
• Designs may be experimental or non-experimental, hypothesis-testing in nature.
• Experimental designs can be informal (before-and-after, after-only, etc.) or formal
(completely randomized, randomized block, Latin square, factorial designs).
• In preparing the design, factors like means of obtaining data, availability of researchers,
time, skills, and costs are considered.

5. Determining Sample Design

• A sample design is a definite plan for obtaining a sample from a given population.
• Population refers to all items under consideration in a field of inquiry.
• A census involves complete enumeration of all items, but it is often costly, time-
consuming, and impractical.
• Hence, a sample is usually taken, which saves time, cost, and effort.
• A sample design specifies how a sample will be chosen before data collection begins.
• Samples may be probability samples (random, stratified, cluster, systematic) or non-
probability samples (purposive, convenience, quota).
• Each technique has its uses: random sampling ensures equal chances, stratified
sampling ensures representation, and quota sampling is less expensive.
• The choice of design depends on nature of the inquiry, accuracy needed, and resources
available.
• Mixed sampling may also be used in practice.
6. Collecting the Data

• Once the sample design is decided, data collection is undertaken.

• Primary data may be collected through observation, interviews, questionnaires, or
schedules.
• Observation provides first-hand data but is costly and limited in scope.
• Personal interviews follow a structured procedure, while telephone interviews are
faster but limited.
• Mailing questionnaires is the most extensively used method in business and economic
surveys.
• In schedules, trained enumerators fill in responses from interviewees.
• Secondary data can be obtained from government reports, publications, books, and
earlier studies.
• The choice of method depends on problem objectives, scope of inquiry, time, finance,
and accuracy required.
• Common sense and experience play a big role in choosing the correct method.

7. Execution of the Project

• Execution is a very important step; faulty execution can make the research useless.
• Data collection must be systematic, adequate, and reliable.
• Interviewers should be properly selected and trained with instruction manuals.
• Field checks should be done to ensure sincerity and efficiency.
• Unanticipated factors must be carefully handled.
• Surveys should be kept under statistical control to maintain accuracy standards.
• For non-response issues, methods like sub-samples of non-respondents may be used.
• Cooperation of respondents is important, and vigorous efforts may be made to secure
it.
• Execution determines the quality of the data collected.

8. Analysis of Data

• After data are collected, analysis begins with editing, coding, and tabulation.
• Editing improves quality, while coding transforms categories into symbols for
tabulation.
• Tabulation arranges data into tables, often with the help of computers in large
inquiries.
 • Analysis involves computation of percentages, averages, coefficients, and applying
statistical tests.
• Data are condensed into manageable groups for meaningful study.
• Significance tests like t-test, chi-square, or variance analysis may be applied.
• Statistical analysis helps determine whether results are valid or due to chance.
• Computers make analysis faster and capable of handling many variables
simultaneously.
• Careful analysis is essential for sound conclusions.

9. Hypothesis Testing

• After analyzing data, hypotheses (if any) are tested.

• Tests such as chi-square, t-test, and F-test are commonly used.
• Testing shows whether the data support or contradict the hypothesis.
• Hypotheses may be accepted, rejected, or modified.
• In absence of initial hypotheses, generalizations may be stated as hypotheses for later
testing.
• Hypothesis testing confirms or disproves assumptions made earlier.
• The process adds validity to the research conclusions.
• It also provides a basis for future studies.
• This step is critical for ensuring the logical strength of findings.

10. Generalizations and Interpretation

• If hypotheses are tested and upheld repeatedly, the researcher may generalize the
findings.
• Generalizations may lead to building of theories.
• Interpretation explains the findings based on theory and logic.
• It is the real value of research as it links results to broader knowledge.
• Interpretation may also raise new questions and suggest further research.
• The process requires logical thinking and unbiased judgment.
• Interpretation often involves relating results to existing knowledge.
• It enables meaningful insights into the problem studied.
 • Without proper interpretation, research results remain meaningless.

11. Preparation of the Report or Thesis

• The last step is preparing the report or thesis.

• It must be written carefully, clearly, and precisely.
• The report layout includes: (i) preliminary pages (title, acknowledgements, contents),
(ii) main text (introduction, findings, report body, conclusion), and (iii) end matter
(appendices, bibliography, index).
• The introduction should define objectives, methodology, scope, and limitations.
• Findings and recommendations should be summarized in simple language.
• The main body must be logically structured and presented in sections.
• Charts, graphs, and tables should be used only if they present data clearly.
• Language must be simple, concise, and free from vague expressions.
• Bibliography and appendices should include all supporting details.